[1/3] rtc: Allow including mc146818 rtc header from userspace

Submitted by Sasha Levin on Nov. 29, 2011, 8:35 a.m.

Details

Message ID 1322555735-32163-1-git-send-email-levinsasha928@gmail.com
State Rejected
Headers show

Commit Message

Sasha Levin Nov. 29, 2011, 8:35 a.m.
This patch moves kernel specific header includes into the kernel #ifdef-ed
section, thus allowing userspace to include this header directly.

Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: rtc-linux@googlegroups.com
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
---
 include/linux/mc146818rtc.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Comments

Sasha Levin Dec. 14, 2011, 8:56 a.m.
Just a ping to RTC maintainers. Any comments on the patch? can it go in?

On Tue, 2011-11-29 at 10:35 +0200, Sasha Levin wrote:
> This patch moves kernel specific header includes into the kernel #ifdef-ed
> section, thus allowing userspace to include this header directly.
> 
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Cc: rtc-linux@googlegroups.com
> Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
> ---
>  include/linux/mc146818rtc.h |    3 +--
>  1 files changed, 1 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/mc146818rtc.h b/include/linux/mc146818rtc.h
> index 2f4e957..49d93ff 100644
> --- a/include/linux/mc146818rtc.h
> +++ b/include/linux/mc146818rtc.h
> @@ -11,11 +11,10 @@
>  #ifndef _MC146818RTC_H
>  #define _MC146818RTC_H
>  
> +#ifdef __KERNEL__
>  #include <asm/io.h>
>  #include <linux/rtc.h>			/* get the user-level API */
>  #include <asm/mc146818rtc.h>		/* register access macros */
> -
> -#ifdef __KERNEL__
>  #include <linux/spinlock.h>		/* spinlock_t */
>  extern spinlock_t rtc_lock;		/* serialize CMOS RAM access */
>
Jean-Christophe PLAGNIOL-VILLARD Dec. 14, 2011, 5:42 p.m.
On 10:56 Wed 14 Dec     , Sasha Levin wrote:
> Just a ping to RTC maintainers. Any comments on the patch? can it go in?

why do you need such ocde in the userspace?

Best Regards,
J.
Cyrill Gorcunov Dec. 14, 2011, 5:55 p.m.
On Wed, Dec 14, 2011 at 06:42:51PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 10:56 Wed 14 Dec     , Sasha Levin wrote:
> > Just a ping to RTC maintainers. Any comments on the patch? can it go in?
> 
> why do you need such ocde in the userspace?
> 

See patch 2/3 in this series. We use this constants in kvmtool, so it's
preferred to not duplicate code if possible.

	Cyrill
Jean-Christophe PLAGNIOL-VILLARD Dec. 18, 2011, 12:10 p.m.
On 21:55 Wed 14 Dec     , Cyrill Gorcunov wrote:
> On Wed, Dec 14, 2011 at 06:42:51PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 10:56 Wed 14 Dec     , Sasha Levin wrote:
> > > Just a ping to RTC maintainers. Any comments on the patch? can it go in?
> > 
> > why do you need such ocde in the userspace?
> > 
> 
> See patch 2/3 in this series. We use this constants in kvmtool, so it's
> preferred to not duplicate code if possible.
why do you need it?

Best Regards,
J.
Cyrill Gorcunov Dec. 18, 2011, 12:42 p.m.
On Sun, Dec 18, 2011 at 01:10:08PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > Just a ping to RTC maintainers. Any comments on the patch? can it go in?
> > > 
> > > why do you need such ocde in the userspace?
> > 
> > See patch 2/3 in this series. We use this constants in kvmtool, so it's
> > preferred to not duplicate code if possible.
>
> why do you need it?
> 

Already replied in previous mail.

	Cyrill
Pekka Enberg Dec. 19, 2011, 7:20 a.m.
> On 21:55 Wed 14 Dec     , Cyrill Gorcunov wrote:
>> On Wed, Dec 14, 2011 at 06:42:51PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
>>> On 10:56 Wed 14 Dec     , Sasha Levin wrote:
>>>> Just a ping to RTC maintainers. Any comments on the patch? can it go in?
>>>
>>> why do you need such ocde in the userspace?
>>
>> See patch 2/3 in this series. We use this constants in kvmtool, so it's
>> preferred to not duplicate code if possible.

On Sun, 18 Dec 2011, Jean-Christophe PLAGNIOL-VILLARD wrote:
> why do you need it?

It's for RTC emulation code in the KVM tool. For the patch:

Acked-by: Pekka Enberg <penberg@kernel.org>

Given that the RTC maintainers are extremely unresponsive, I'm CC'ing 
Andrew too.

 			Pekka
Jean-Christophe PLAGNIOL-VILLARD Dec. 20, 2011, 2:27 p.m.
On 16:42 Sun 18 Dec     , Cyrill Gorcunov wrote:
> On Sun, Dec 18, 2011 at 01:10:08PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > > Just a ping to RTC maintainers. Any comments on the patch? can it go in?
> > > > 
> > > > why do you need such ocde in the userspace?
> > > 
> > > See patch 2/3 in this series. We use this constants in kvmtool, so it's
> > > preferred to not duplicate code if possible.
> >
> > why do you need it?
> > 
> 
> Already replied in previous mail.
still not clear why you do not use the RTC API

sorry this need to be more clear

Best Regards,
J.
Cyrill Gorcunov Dec. 20, 2011, 2:36 p.m.
On Tue, Dec 20, 2011 at 03:27:39PM +0100, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > 
> > Already replied in previous mail.
>
> still not clear why you do not use the RTC API
> sorry this need to be more clear
> 

Jean, I might be missing somethig, which API you mean?
Since we emulate RTC functions we need those constants
in user-space app so having them just in some common
header (in this case -- kernel header) would be pretty
convenient -- we would simply grab them and would not
duplicate code (by having our own header with the same
constants). Sounds good?

	Cyrill

Patch hide | download patch | download mbox

diff --git a/include/linux/mc146818rtc.h b/include/linux/mc146818rtc.h
index 2f4e957..49d93ff 100644
--- a/include/linux/mc146818rtc.h
+++ b/include/linux/mc146818rtc.h
@@ -11,11 +11,10 @@ 
 #ifndef _MC146818RTC_H
 #define _MC146818RTC_H
 
+#ifdef __KERNEL__
 #include <asm/io.h>
 #include <linux/rtc.h>			/* get the user-level API */
 #include <asm/mc146818rtc.h>		/* register access macros */
-
-#ifdef __KERNEL__
 #include <linux/spinlock.h>		/* spinlock_t */
 extern spinlock_t rtc_lock;		/* serialize CMOS RAM access */